A permutation-based approach to inference for weighted sum regression with correlated chemical mixtures.
Ontology highlight
ABSTRACT:
SUBMITTER: Lyden GR
PROVIDER: S-EPMC9883011 | biostudies-literature | 2022 Apr
REPOSITORIES: biostudies-literature
Publications
A permutation-based approach to inference for weighted sum regression with correlated chemical mixtures.
Lyden Grace R GR Vock David M DM Barrett Emily S ES Sathyanarayana Sheela S Swan Shanna H SH Nguyen Ruby Hn RH
Statistical methods in medical research 20220206 4
There is a growing demand for methods to determine the effects that chemical mixtures have on human health. One statistical challenge is identifying true "bad actors" from a mixture of highly correlated predictors, a setting in which standard approaches such as linear regression become highly variable. Weighted Quantile Sum regression has been proposed to address this problem, through a two-step process where mixture component weights are estimated using bootstrap aggregation in a training datas ...[more]
Similar Datasets
Project description:Individuals are exposed to a large number of diverse environmental chemicals simultaneously and the evaluation of multiple chemical exposures is important for identifying cancer risk factors. The measurement of a large number of chemicals (the exposome) in epidemiologic studies is allowing for a more comprehensive assessment of cancer risk factors than was done in earlier studies that focused on only a few chemicals. Empirical evidence from epidemiologic studies shows that chemicals from different chemical classes have different magnitudes and directions of association with cancers. Given increasing data availability, there is a need for the development and assessment of statistical methods to model environmental cancer risk that considers a large number of diverse chemicals with different effects for different chemical classes. The method of grouped weighted quantile sum (GWQS) regression allows for multiple groups of chemicals to be considered in the model such that different magnitudes and directions of associations are possible for each group of chemicals. In this paper, we assessed the ability of GWQS regression to estimate exposure effects for multiple chemical groups and correctly identify important chemicals in each group using a simulation study. We compared the performance of GWQS regression with WQS regression, the least absolute shrinkage and selection operator (lasso), and the group lasso in estimating exposure effects and identifying important chemicals. The simulation study results demonstrate that GWQS is an effective method for modeling exposure to multiple groups of chemicals and compares favorably with other methods used in mixture analysis. As an application, we used GWQS regression in the California Childhood Leukemia Study (CCLS), a population-based case-control study of childhood leukemia in California to estimate exposure effects for many chemical classes while also adjusting for demographic factors. The CCLS analysis found evidence of a positive association between exposure to the herbicide dacthal and an increased risk of childhood leukemia.
| S-EPMC7827322 | biostudies-literature
Project description:In risk evaluation, the effect of mixtures of environmental chemicals on a common adverse outcome is of interest. However, due to the high dimensionality and inherent correlations among chemicals that occur together, the traditional methods (e.g. ordinary or logistic regression) suffer from collinearity and variance inflation, and shrinkage methods have limitations in selecting among correlated components. We propose a weighted quantile sum (WQS) approach to estimating a body burden index, which identifies "bad actors" in a set of highly correlated environmental chemicals. We evaluate and characterize the accuracy of WQS regression in variable selection through extensive simulation studies through sensitivity and specificity (i.e., ability of the WQS method to select the bad actors correctly and not incorrect ones). We demonstrate the improvement in accuracy this method provides over traditional ordinary regression and shrinkage methods (lasso, adaptive lasso, and elastic net). Results from simulations demonstrate that WQS regression is accurate under some environmentally relevant conditions, but its accuracy decreases for a fixed correlation pattern as the association with a response variable diminishes. Nonzero weights (i.e., weights exceeding a selection threshold parameter) may be used to identify bad actors; however, components within a cluster of highly correlated active components tend to have lower weights, with the sum of their weights representative of the set.
| S-EPMC6261506 | biostudies-literature
Project description:BackgroundOptimization of mixture analyses is critical to assess potential impacts of human exposures to multiple pollutants. Single-index regression methods quantify total mixture association and chemical component contributions. Single-index methods include several variants of quantile g-computation (QGC) and weighted quantile sum regression (WQSr), though each has limitations.ObjectivesWe developed a novel permutation test for WQSr and compared its performance to extant versions of WQSr and QGC in simulation studies and an analysis of prenatal phthalates and childhood cognition.MethodsWQSr uses ensemble nonlinear optimization to identify weights for mixture exposures in an index associated with the outcome in a prespecified direction, with ensembles based on bootstrap resampling (WQSBS) or random subsetting of exposures (WQSRS). Statistical significance can be assessed without splitting the data (Nosplit), by splitting the data into training and test sets (Split), by repeatedly holding out test sets (RH), or by using a novel permutation test (PT) to obtain a more accurate p-value. QGC instead provides a sum mixture coefficient and component coefficients with no constraints on direction. In simulations, we compared false positive rates (FPR) and power to detect true associations and accuracy in estimating mixture weights. We also estimated associations between prenatal phthalate mixtures and childhood IQ in the Conditions Affecting Neurocognitive Development and Learning in Early Childhood cohort using each method.ResultsFPR was well controlled at ≤7% in all but the Nosplit WQSr variants. Among these methods, the WQSBS and WQSRS PT variants had the highest power (89%-98%), with lower power for QGC (85%-93%) and RH (60%-97%) or Split WQSr variants (40%-78%). WQSr methods estimated mixture weights 2-4 times more accurately than the QGC method. Coefficients for mixture associations with full scale IQ varied 3- to 4-fold across analytic methods.DiscussionWQSr paired with our novel permutation test best balanced power and false positive rate when assessing a mixture effect. As new methods develop, each should be examined for performance and applicability. https://doi.org/10.1289/EHP10570.
| S-EPMC9426671 | biostudies-literature
Project description:Studies of the health effects of the microbiome often measure overall associations by using diversity metrics, and individual taxa associations in separate analyses, but do not consider the correlated relationships between taxa in the microbiome. In this study, we applied random subset weighted quantile sum regression with repeated holdouts (WQSRSRH), a mixture method successfully applied to 'omic data to account for relationships between many predictors, to processed amplicon sequencing data from the Human Microbiome Project. We simulated a binary variable associated with 20 operational taxonomic units (OTUs). WQSRSRH was used to test for the association between the microbiome and the simulated variable, adjusted for sex, and sensitivity and specificity were calculated. The WQSRSRH method was also compared to other standard methods for microbiome analysis. The method was further illustrated using real data from the Growth and Obesity Cohort in Chile to assess the association between the gut microbiome and body mass index. In the analysis with simulated data, WQSRSRH predicted the correct directionality of association between the microbiome and the simulated variable, with an average sensitivity and specificity of 75% and 70%, respectively, in identifying the 20 associated OTUs. WQSRSRH performed better than all other comparison methods. In the illustration analysis of the gut microbiome and obesity, the WQSRSRH analysis identified an inverse association between body mass index and the gut microbe mixture, identifying Bacteroides, Clostridium, Prevotella, and Ruminococcus as important genera in the negative association. The application of WQSRSRH to the microbiome allows for analysis of the mixture effect of all the taxa in the microbiome, while simultaneously identifying the most important to the mixture, and allowing for covariate adjustment. It outperformed other methods when using simulated data, and in analysis with real data found results consistent with other study findings.
| S-EPMC9819204 | biostudies-literature
Project description:Detecting the temporally and spatially varying correlations is important to understand the biological and disease systems. Here we proposed a geographically weighted temporally correlated logistic regression (GWTCLR) model to identify such dynamic correlation of predictors on binomial outcome data, by incorporating spatial and temporal information for joint inference. The local likelihood method is adopted to estimate the spatial relationship, while the smoothing method is employed to estimate the temporal variation. We present the construction and implementation of GWTCLR and the study of the asymptotic properties of the proposed estimator. Simulation studies were conducted to evaluate the robustness of the proposed model. GWTCLR was applied on real epidemiologic data to study the climatic determinants of human seasonal influenza epidemics. Our method obtained results largely consistent with previous studies but also revealed certain spatial and temporal varying patterns that were unobservable by previous models and methods.
| S-EPMC5780421 | biostudies-literature
Project description:BackgroundNew statistical methodologies were developed in the last decade to face the challenges of estimating the effects of exposure to multiple chemicals. Weighted Quantile Sum (WQS) regression is a recent statistical method that allows estimating a mixture effect associated with a specific health effect and identifying the components that characterize the mixture effect.ObjectivesIn this study, we propose an extension of WQS regression that estimates two mixture effects of chemicals on a health outcome in the same model through the inclusion of two indices, one in the positive direction and one in the negative direction, with the introduction of a penalization term.MethodsTo evaluate the performance of this new model we performed both a simulation study and a real case study where we assessed the effects of nutrients on obesity among adults using the National Health and Nutrition Examination Survey (NHANES) data.ResultsThe method showed good performance in estimating both the regression parameter and the weights associated with the single elements when the penalized term was set equal to the magnitude of the Akaike information criterion of the unpenalized WQS regression. The two indices further helped to give a better estimate of the parameters [Positive direction Median Error (PME): 0.022; Negative direction Median Error (NME): -0.044] compared to the standard WQS without the penalization term (PME: -0.227; NME: 0.215). In the case study, WQS with two indices was able to find a significant effect of nutrients on obesity in both directions identifying sodium and magnesium as the main actors in the positive and negative association, respectively.DiscussionThrough this work, we introduced an extension of WQS regression that improved the accuracy of the parameter estimates when considering a mixture of elements that can have both a protective and a harmful effect on the outcome; and the advantage of adding a penalization term when estimating the weights.
| S-EPMC10392701 | biostudies-literature
Project description:Chemical-protein interaction (CPI) is the central topic of target identification and drug discovery. However, large scale determination of CPI is a big challenge for in vitro or in vivo experiments, while in silico prediction shows great advantages due to low cost and high accuracy. On the basis of our previous drug-target interaction prediction via network-based inference (NBI) method, we further developed node- and edge-weighted NBI methods for CPI prediction here. Two comprehensive CPI bipartite networks extracted from ChEMBL database were used to evaluate the methods, one containing 17,111 CPI pairs between 4,741 compounds and 97 G protein-coupled receptors, the other including 13,648 CPI pairs between 2,827 compounds and 206 kinases. The range of the area under receiver operating characteristic curves was 0.73 to 0.83 for the external validation sets, which confirmed the reliability of the prediction. The weak-interaction hypothesis in CPI network was identified by the edge-weighted NBI method. Moreover, to validate the methods, several candidate targets were predicted for five approved drugs, namely imatinib, dasatinib, sertindole, olanzapine and ziprasidone. The molecular hypotheses and experimental evidence for these predictions were further provided. These results confirmed that our methods have potential values in understanding molecular basis of drug polypharmacology and would be helpful for drug repositioning.
| S-EPMC3397956 | biostudies-literature
Project description:BackgroundPersons with a college degree are more likely to engage in eHealth behaviors than persons without a college degree, compounding the health disadvantages of undereducated groups in the United States. However, the extent to which quality of recent eHealth experience reduces the education-based eHealth gap is unexplored.ObjectiveThe goal of this study was to examine how eHealth information search experience moderates the relationship between college education and eHealth behaviors.MethodsBased on a nationally representative sample of adults who reported using the Internet to conduct the most recent health information search (n=1458), I evaluated eHealth search experience in relation to the likelihood of engaging in different eHealth behaviors. I examined whether Internet health information search experience reduces the eHealth behavior gaps among college-educated and noncollege-educated adults. Weighted logistic regression models were used to estimate the probability of different eHealth behaviors.ResultsCollege education was significantly positively related to the likelihood of 4 eHealth behaviors. In general, eHealth search experience was negatively associated with health care behaviors, health information-seeking behaviors, and user-generated or content sharing behaviors after accounting for other covariates. Whereas Internet health information search experience has narrowed the education gap in terms of likelihood of using email or Internet to communicate with a doctor or health care provider and likelihood of using a website to manage diet, weight, or health, it has widened the education gap in the instances of searching for health information for oneself, searching for health information for someone else, and downloading health information on a mobile device.ConclusionThe relationship between college education and eHealth behaviors is moderated by Internet health information search experience in different ways depending on the type of eHealth behavior. After controlling for college education, it was found that persons who experienced more fruitful Internet health information searches are generally less likely to engage in eHealth behaviors.
| S-EPMC5081480 | biostudies-literature
Project description:Permutation tests are increasingly being used as a reliable method for inference in neuroimaging analysis. However, they are computationally intensive. For small, non-imaging datasets, recomputing a model thousands of times is seldom a problem, but for large, complex models this can be prohibitively slow, even with the availability of inexpensive computing power. Here we exploit properties of statistics used with the general linear model (GLM) and their distributions to obtain accelerations irrespective of generic software or hardware improvements. We compare the following approaches: (i) performing a small number of permutations; (ii) estimating the p-value as a parameter of a negative binomial distribution; (iii) fitting a generalised Pareto distribution to the tail of the permutation distribution; (iv) computing p-values based on the expected moments of the permutation distribution, approximated from a gamma distribution; (v) direct fitting of a gamma distribution to the empirical permutation distribution; and (vi) permuting a reduced number of voxels, with completion of the remainder using low rank matrix theory. Using synthetic data we assessed the different methods in terms of their error rates, power, agreement with a reference result, and the risk of taking a different decision regarding the rejection of the null hypotheses (known as the resampling risk). We also conducted a re-analysis of a voxel-based morphometry study as a real-data example. All methods yielded exact error rates. Likewise, power was similar across methods. Resampling risk was higher for methods (i), (iii) and (v). For comparable resampling risks, the method in which no permutations are done (iv) was the absolute fastest. All methods produced visually similar maps for the real data, with stronger effects being detected in the family-wise error rate corrected maps by (iii) and (v), and generally similar to the results seen in the reference set. Overall, for uncorrected p-values, method (iv) was found the best as long as symmetric errors can be assumed. In all other settings, including for familywise error corrected p-values, we recommend the tail approximation (iii). The methods considered are freely available in the tool PALM - Permutation Analysis of Linear Models.
| S-EPMC5035139 | biostudies-literature
Project description:There has been a growing interest in the literature on multiple environmental risk factors for diseases and an increasing emphasis on assessing multiple environmental exposures simultaneously in epidemiologic studies of cancer. One method used to analyze exposure to multiple chemical exposures is weighted quantile sum (WQS) regression. While WQS regression has been demonstrated to have good sensitivity and specificity when identifying important exposures, it has limitations including a two-step model fitting process that decreases power and model stability and a requirement that all exposures in the weighted index have associations in the same direction with the outcome, which is not realistic when chemicals in different classes have different directions and magnitude of association with a health outcome. Grouped WQS (GWQS) was proposed to allow for multiple groups of chemicals in the model where different magnitude and direction of associations are possible for each group. However, GWQS shares the limitation of WQS of a two-step estimation process and splitting of data into training and validation sets. In this paper, we propose a Bayesian group index model to avoid the estimation limitation of GWQS while having multiple exposure indices in the model. To evaluate the performance of the Bayesian group index model, we conducted a simulation study with several different exposure scenarios. We also applied the Bayesian group index method to analyze childhood leukemia risk in the California Childhood Leukemia Study (CCLS). The results showed that the Bayesian group index model had slightly better power for exposure effects and specificity and sensitivity in identifying important chemical exposure components compared with the existing frequentist method, particularly for small sample sizes. In the application to the CCLS, we found a significant negative association for insecticides, with the most important chemical being carbaryl. In addition, for children who were born and raised in the home where dust samples were taken, there was a significant positive association for herbicides with dacthal being the most important exposure. In conclusion, our approach of the Bayesian group index model appears able to make a substantial contribution to the field of environmental epidemiology.
| S-EPMC8037139 | biostudies-literature